Skelerose - Angel

More Posts from Skelerose and Others

skelerose

5 months ago

Tags

my art artists on tumblr sketch oc doodle

6 notes View Post

skelerose

1 year ago

52 notes View Post

skelerose

10 months ago

“Omg dungeons menshi is so good! The characters and the worldbuilding and the art and the hu—“

It’s because the author had TWO MONTHS between publications instead of the usual week. 8X the amount of time to plot her story and see where she had been and where she was going and figure out how best to get there. Imagine if this was the industry standard instead of the crushing gears of weekly publication which grinds the minds, bodys, and souls of mangaka to dust until there is nothing left. Imagine Dungeon Menshi qualify manga as a rule not the exception. Imagine stories that did not destroy their creators. Imagine—

Tags

dungeon meshi

69K notes View Post

skelerose

3 months ago

Tags

furry poll snail snails

3 notes View Post

skelerose

1 year ago

top surgery goals

174 notes View Post

skelerose

11 months ago

AXIS 《1995》 Mermaid (Unlicensed Doujin Game)

AXIS 《1995》 Mermaid (Unlicensed Doujin Game)

Tags

pixel art sailor moon

1K notes View Post

skelerose

1 year ago

You can do it! (Throwing a little tantrum along the way never hurt)

This is a special print for patrons this month! ✨

36K notes View Post

skelerose

1 year ago

[みさいる]

Adachi Rei is a crowdfunded life-sized robot project started by missile39 in 2017. She was given a computer-generated voice using the UTAU program and later used for a text-to-speech software called REPLIVOICE Adachi Rei, powered by the A.I. VOICE engine.

メカニカルガール

等身大美少女ロボット開発

80 notes View Post

skelerose

1 year ago

A long infographic with visual aids starting with the conversation:
"Is Miku AI?"
"No."
"Are vocal synths ethical?"
"Yes."
"How so?"
First section is Compensation.
Hatsune Miku is made out of recordings by Saki Fujita. Saki Fujita is contracted to record Miku samples, and is paid for her work.

Section: Recording Method.
This is why Miku is not AI: Saki Fujita records from a list of sounds. It's necessary to have at least one recording per sound Miku should be able to sing. (Visual aid has examples of these sounds, such as "kaka".)
She can also sing the recording list a second time in a different octave, so that she sounds more natural.
Section: Labelling.
The samples Saki Fujita sung are then labelled with what sound they make. These sounds are then reproduced by the engine. This is how Vocal Synth software such as VOCALOID and UTAU work. This model is called "concatenative". (Visual aid shows how "kaka" is split into "k" and "a", which is how it looks in the VOCALOID software.)

Section: User interfacing.
These voicebanks are very flat. Users must adjust the vocals themselves in order to produce singing. This is referred to as "tuning". If you listen to "Tuning BLANK in the style of Vocaloid producers", you can see there are countless ways to tune Hatsune Miku. It is considered a form of artistic expression.
Compare Scratchin' Melodii's original songs to the updated versions. This is the result of hiring an experienced Vocaloid tuner.
Question: How do AI Vocal Synths work?
Answer: They are actually extremely similar!

Section: Compensation.
Let's use the Synthesizer V Studio library "Solaria". Solaria is made out of recordings by Emma Rowley. Emma Rowley is contracted to record Solaria samples, and paid for her work.
Section: Recording.
Emma Rowley then records several hours of singing data. This is the substance of the library.
Section: Base model.
The AI needs a base to understand what it's interpreting. Unlike images, there is a large amount of volunteer voice data out there. It's typically assumed that base models are trained ethically. (Visual aid shows Dreamtonics, the developer company behind Synthesizer V, asking a university "Can I use this voice data you made for TTS research?" and observing a person saying "Hi! Here is a few hours of singing data you can use for voice technology.")

Section: Labelling.
Labelling is also the same. The singing is broken up into phonemes the engine will interpret.
Header Section: Deep Learning.
In casual speech, "AI" refers to computer learning/sorting algorithms. "Diffusion" AI is the result of DNN; Deep Neural Network. It is the most drastic difference between concatenative and AI voicebanks.
Section: Teaching the base model.
The computer must be taught what the sounds are. The concept it builds is the "base model". (Visual guide is a cartoon of two computers talking. "Here's a british man saying 'bath'." "Added to my concept of 'a'." "Here's a Japanese girl saying 'baka'." "Added to my concept of 'a'.")
Section: Training the voice model.
Emma Rowley's recordings are then made into a reference point. This will make it so it will only render based on what it knows about Emma Rowley's singing. (Visual aid is a similar cartoon where a person talks to a computer while giving it a drive. Computer: "Now that I know what 'a' is, how should it sound?" Person: "I've labelled every time Emma Rowley says 'a'. Use this!")

Section: Diffusion.
The Solaria model uses everything it learned from Emma Rowley's recordings and the base mdoel to determine how 'a' sounds based on what note it's sung on, what's next to it, etcetera.
Section: Interfacing.
Tuners have been mixed on this; it sounds much clearer, yet the AI also has voice pitch models, so there's not as much as an incentive to develop your own personal flair.
Question: Are voice changers ethical?
Answer: Oh geez.

Section: ARE they ethical?
We don't need to break this down a third time. Voice changers are the generative AI of voice synthesis. It requires a lot less work of both the developer and the user, a simple applicator of everything the machine knows onto a piece of audio. What are the ranges of ethics?
Vocaloid 6 is packaged with a voice changer. It is only for AI libraries, voiced by people who agreed to this and were compensated. This is definitely ethical.
If you bought Hatsune Miku, you're nominally permitted to use the results as you see fit. Is tuning Miku and then creating a voice changer of her singing ethical? I genuinely don't know.
There's also a question of art. If you were to project the voice actor onto your own personal tuning work, isn't that still artistic expression? A voice is different from an art style. Where is human expression being interrupted by automation? I can't make an explainer for those subjective concepts.
I hope you're now educated enough to think on it yourself. End of image transcription.

A lot of people try to explain this without knowing anything about how voice synthesis works, so here's my breakdown on No, Hatsune Miku Is Not AI, And No, AI Voice Synthesis Is Not Bad.

20K notes View Post

skelerose

1 year ago